Garden City Ruby 2014 - Lightning Talks

SPEAKER 1: My name is Smith. I work at Flipkart.

SPEAKER 2: Hi guys. I'm Tri Hari??.
I

am a partner at Nilenso.

S.1.: OK, we are going to talk about impact.

So what do we mean?

So let's start with a question.

As programmer, what are the problem we face?
So

looking at this, aside from that, you know,
maybe,

in the worst case we have to quickly fix

a bug on production, or maybe just really
cranky

client comes up with a requirement which is
very

hard to fulfill, or you know, our code quality

or our tests written so big, like Chad said

yesterday, that it takes a long time to finish.

But these are the kind of problems we face,

generally. But let's see what other problems
do people

in rural India face?

Yeah. So. Let's start with this: food.

So I'm sure all of us enjoyed the really

good lunch there, [00:01:25] pune gosh, chicken
packenard, fune

mokney [00:01:28]. We will also get a tea
break

soon enough. I'm hoping. But what about the
people

out there? They don't even get a square meal

once a day.

Water- well. People in India really have accepted
the

fact that they don't get water to bath in

the rural areas. They somehow manage it with
the

drinking water they get. It's really not clean,
but

they somehow manage.

Shelter. V.J. told me a fact today, that sixty

percent of population in India do not have
a

toilet. They don't have access to a toilet.
So

let's, not even care about anything else.

And education, this is something very personal
to me.

Where I come from, it's even lucky to people

to study up to ?? standard [00:02:12].

So, so my question is, how many of you

have written a line of code which has impacted

life in some way? I'm not saying a big

way, but a meaningful way.

Can anyone show their hands? Yeah.

And how many of you really want to help

them out.

Only this many? Like this is really sad. OK.

So you really want to help them out, and

this is how you can do it.

S.2.: Right, so this is essentially what we
heard

about. A few of us at Nilenso, and previously

C42, have been fortunate enough to work on
the

Survey Project.

And I'm gonna leave it to this: We have

6,000 people who have been impacted by the
first

run of this project. This thing is opensource.
I'm

gonna leave this slide for you guys to look

at. And we've- five big cities, 6,000 people.
Kids,

pregnant women, with the nutrition survey.
Right, this was

the first run.

And there was another one, that went to people,

women, in the agricultural side, and there
are three

thousand families who have been helped. And
one, a

single fact that, that's gonna help you understand
where

we are with this - look at the last

three lines in this slide, basically.

These are the inferences that have been drawn
based

on the data here that is 6,000 requests from

here, and 3,000 families. That's roughly,
again, around 6,000

records here, right.

So that's so much data coming in, and inferences

drawn across surveys.

Technically, like, why am I, why are we presenting

this here, and what do you people have to

do with it? Apart from this, like, the technical,

like the number of apps and whatnot, there
is

so much data coming in, right. The last thing

that we have in our road map is a

separate Clojure app to take care of the big

data stuff.

And, like, personally I have spent hours of
my

time looking at the UX in this thing.

So these apps have to go down to the

last mile, down to people's homes, where there's
no

internet, right. And they have, like, one
hour of

internet in a week, and then there's so much

data integrated problems that you have to
deal with

as well.

And sadly over the last month or so, this

has come to a slow stop, and we have

not had the time to work on this. But,

I'm hoping that a lot of people here do

have some time to spare. Send us a poll

request, whatever it is.

These are the people we really care about.
We

write algorithms to bring down our performance
from point

zero one seconds to point zero zero one seconds,

and that's all we have, right.

S.1.: So I'll wrap it up really fast. So

let's not always write code to help these
people

out with our wonderful projects like WunderRails
RubyGems and

other languages.

Let's also these people out as well. You know,

maybe give them a boost, because I really
believe

we don't all have to be Bill Gates to

help these people out. With our own skills
we

can contribute to and write technical projects
that could

help them out.

And, yeah, this is where you can find the

projects and start and everything.

Thank you.

S.2.: You can catch anyone of us after the

talk if you want to help us out. Thanks.

INTRO: So next up we have Kestap?? [00:06:12]
on

the stage. Time starts now.

SPEAKER: Hi everyone. Today I'll be talking
about Sinatra,

basic introduction. Probably everybody knows
this, but more and

more though I'll be talking about how to write

an extension in Sinatra.

So the basic structure of Sinatra is this.
There

is a request class that extends from Rack::Request,
response

that extends from Rack::Response, ExtendedRack
is basically for the

things that are not in rack, and so they

define extra methods there.

There is a commonlogger which extends from
Rack::CommonLogger, and

this is the main part, Base.

So there are actually three ways to, there
are

actually three ways to extending Sinatra.
The first one

is extensions, the second one is - no.

Not quite.

The second one is helpers and the third one

is middleware. I won't be talking about middleware
probably,

there is a lot of material on that. but

the extensions and helpers, I've found it
difficult to

use them, so. I've kind of made a small

presentation here.

So the first difference between helpers and
extensions is

that, in helpers, you use a helper or you

write a helper if the functionality is required
in

the get method or post method or any other

blocks that we use.

So inside a route block the extensions are
for

extending the functionality of Sinatra itself.

So for example, inside a get block we find

our two, return JSON response every time,
so I

can make a function like this, JSON(a) and
convert

basically converts our hash into a JSON string.

So the implementation is this. You write it
in

a module called Helpers. You define the method,
you

pass it in, and it basically returns the string.

And you include your helpers in before, in
your

app file.

So to write this as a gem you have

to put that under a namespace of Sinatra,
then

use the module, the same code as above.

And these two are the magic things that actually

happen. So base dot helpers is almost similar
to

adding helpers json inside your app file.

So, and base dot set is basically, it sets

the json_content_type, sets the request_content_type
to JSON by default

for every request.

So the other thing, the major portion is extensions.

So in extensions, Sinatra provides extensions
to, let's say

you want to define a new route, so you

can do that by extensions. I'll demonstrate
a small

thing, a small route actually. So I created
this

extension today.

So I have a module Sinatra, I have this

super module, and this is an extra route.
So

this, the route method is provided by Sinatra
base,

and I'm extending that, so that every, every
request

that you make to this end point, to, with

a method request of this one, instead of get

r post, it will return something pretty good.

So start it up.

And yeah. So that's what it does.

So, you don't have to clap for that, it's

lame.

So the basic difference between these two
is that

this is, this is an extension of Sinatra.
This

route is not defined but we are defining that

route.

We are using the things that are provided
by

Sinatra and we are basically extending. So
the register

method is a method provided by Sinatra where
all,

whatever module you define have to be registered
with

Sinatra, so that every app that you use, and,

like this. So every app that you use, you

just call register Sinatra, that one, and
I'll, I've

defined my route instead of get slash, I have

used doge and it returns this.

So the basic difference is that it's, like
you,

if you want to extend so there are these

two things, and I think that's it.

VINASH: Yeah, so. This Billard, I am Vinash.
We

are from Mumbai, RV?? [00:10:57] So I just
want

to tell all of you guys, and introduce us

too, we're from Mumbai RV?? [00:11:01].

We have total counted seven meet ups in all.

So we are on our way to have a

big meet up soon. Maybe in June or later.

But till now we have a frequent audience of

ten to twenty people, always there. We have
one

or two talks, we have a single hack session

once. That was for Brails?? [00:11:19]

We are not going to push anything, but we

are able to discuss a few things and we

wrote entire cookies, et cetera et al. Everyone
knows.

So there is a mindset about Bombai that Bombai

is partially, Bombai has very few jobs and
all

that stuff. But I can say that nowadays with

so many startups starting up, the salaries
are very

good I guess, currently.

So I guess I will tell you that. You

should ship to Bombai if you are born in

part of Bombai, you should definitely come
back to

Bombai rather than going to Bangalore and
Pume, because,

yeah.

I know! The intellectual level at Bombai is
a

little low as compared to Bangalore or Pume,
but

we need to change that, and we want guys

like you to come back to Bombai who were

born in a part of Bombai and increase the

?? [00:12:18] of Bombai to a level which can

compete with Bangalore and Pume itself.

So I want Billard to say a few words

about the same.

BILLARD: Actually I just interacted with some
of the

peers here, and I found out that they didn't

even know that there's are some developers
in Bombai

like who know Rails or something. So we just

came here to give a lightning talk that, oh,

dude we are, we exist there, there are some

people who are trying to make a dent, not

in the, not in the universe, but at least

in Bombai.

To, so we are that's what we are putting

all our efforts, like, and that's what, as
my-

as Vanash said that we are about to complete

first annual city for Mombai. I RV?? [00:13:11].
That's

what we are now making the name out of

it.

So we are created at least a small event,

which, so, which we all are planning, looking
forward

to. So that's all we have. And Bombai RV

is alive and it will be.

SATYA KALLURI: Hey guys, this is Satya, I'm
from

Hadribad?? [00:13:36], and you know, I basically
want to

talk about technology meetups. I've been part
of some

technology meetups for quite some time. And
you know

to be very straight, I was a little upset

with the way they go.

So, and fortunately, you know, I've been in
touch

with some of the technologies these days,
and they've

revived the hope in me, so I just wanted

to share some of my learnings, and how can

we make better use of these technology meetups.

Oh, by the way, I'm Satya Kalluri and our

group is @HydRuby, so you can follow us there.

So the first thing is I believe there are

two kinds of meetups in the universe. There
are

two really kind of different meetups.

A) is technology meetups, which we need to
consider

them as absolutely local. It's like you and
me

catch up together in the next house or the

next coffee house. It doesn't have to be as

big as this in total area?? [00:14:21]. So
it's

a technology meetup where you know what, it's
a

local thing. And you know what, the, and it's

focused. It doesn't have to be as wide agenda

as this.

It could be as simple as a bug-a-thon. It

could be as simple as learning a new gem,

or it could be anything basically. A small
local,

focused.

And there are other ones which are technology
conferences.

These are bigger ones, where in what, all,
it's

like one of the organizers said, you know
like,

made of communities and sub communities. Atoms
and molecules,

right.

So these small technology meetups, they come
up together

and create these bigger conferences, regional
ones, which have

a wider agenda.

So my problem is about the technology meetups.
The

conferences are going good. GCRC here is an
example,

they're doing an awesome job. But how can
we

improve the technology meetups. You know whenever
I catch

up with folks at the lunch place or something,

the majority of them say, oh, you know, my

meetup gets only two attendees, three attendees.

You know that over a period of time I've

just found that certain items, if, which if
we

address them properly I think the local technology
meetups

are gonna be kickass.

So.

So. Basically I wanted to let you know what

a technology meetup is, first of all, and
why

are they for?

Technology meetups are absolutely for techies
to collaborate, learn,

contribute, share knowledge. So these are
local ones. So

do understand, these are the only for those
techies.

They're not for the job seekers and not the

employers. Whenever there's a technology meetup
going, and if

you need a Rails developer, please do not
go

there.

And if you're looking for a job, don't go

there. Just, let's not pollute these small
local meetups.

They have for technology peers to basically
go there,

share knowledge and solve each others problems
and become

better in technology.

You know the last two are acceptable for conferences

like these, where, you know, these are big
things

where there are networking opportunities and
tech meetups and

everything. But for local meetups let's stick
to technology.

That way everybody who comes, you know what,
knows

what's going to happen. It's technology. This
is Rails,

which I love. I come here to write, to

learn Rails.

And while participating, you know what, these,
the following

six things are the most important things which
I

could recommend everybody to follow.

The first thing is identity. Most of those
whom

I see don't have an identity on Twitter, don't

have an identity on StackerWorkflow, even
meetup I see

some of them say plus five. Now what the

heck does plus five mean?

Not what, who are the other five guys? So

it would be nice if they identities, because
that

way I might want to communicate with that
fifth

guy. I don't need to contact this first guy.

So and you are gonna create your identity,
somebody

needs to contact you directly, you know what,
it

creates an open channel, and as one of them,

one of the speakers said, you know, you are

going to create and identity and then a brand.

So this is the first step. So create an

identity for yourself on Twitter, StackerWorkflow
and everything.

It adds authenticity to the group, you know.
Two

guys, two persons catch up in a meetup doesn't

make sense. But two plus five is seven, so

when seven people really catch up in a meetup,

I think it adds a lot of authenticity to

the group.

When you guys have your own facebook account,
why

not have your own StackOverflow account. I
don't get

it.

The, so the next thing is RSVPs. I see

many RSVPs, like even our @HydRuby has thirty-five
RSVPs,

but only four of them turned up, so. Take

your RSVPs seriously because there are people,
like, you

know, they're not awake, and you know all
these

people who take these meetups very seriously,
and there

may not be some, some people who buy beers

for everybody who come up and-

When 31 beers are leftover, you know, nobody's
gonna

buy them, so. Please take your RSVP seriously.
That

adds authenticity to the group, OK.

And the next thing is attendees, you know,
don't

wait for ten or twenty people to really come

to a meetup. You just need one other guy

to listen to you and you can create a

meetup, and you can, guys, can really discuss
about

technology.

The next thing is StackOverflow, you know,
just share

and contribute. You don't have to really,
I mean,

for you to really become a good meetup guy,

I think at least spend every day, fifteen
minutes

on StackOverflow. Don't just find answers
but answer the

questions. This I've learned from Swanand.
This guy has

some 6,000 bloody points on StackOverflow.
I have some

twelve hundred, that's it.

So I've learned from it. So become a better

contributer. You become really confident,
and you'll become a

different person. You can also contribute
at GitHub. If

you don't know which GitHub module to really
contribute,

go for Codetriage. It gives you good popular
libraries

to really help.

And, do you have an idea? Just stop working

on it. First learn better technology and then,
you

know, you can definitely work on the product.
And

the last thing is, like, spread the word.
This

guy's already on me, so. Spread the word.
Go

for Facebook. You can use Facebook too, and
you

can also use Twitter.

So spread the word and make these technology
meetups

a good thing.

Cheers guys, thank you.

SPEAKER: Before I start, big thumbs up for
the

guys who started off. It's great to see, you

know, technology people, developers who care
about social issues.

And here I am to talk about Milaap, which

is also a social enterprise. And I just want

to talk about what Milaap is all about and

how you can help.

So Milaap is a Crowdfunding platform for communities
in

need which is powered by Ruby on Rails. So

we are a hardcore Ruby on Rails shop, and

we bridge the gap between people who are need

of help and people who want to help, right.

So let's quickly look at what that means.
So

I'm sure all of us want to do good.

We want to help others. But when it comes

to helping others with financial aid a lot
of

us are hesitant, because we are not sure,
you

know, our hard-earned money, when we give
it to

someone, you know, is it going to be used

in the right way? Is it the right person

I'm going to give this money to.

So Milaap tries to bridge this gap, right.
So

Milaap has a set of profiles uploaded on our

website, which are across various kind of
causes, and

you can, you can filter through various criteria
and

choose them. And it helps you make an informed

decision about whom you want to help, right.

So this is how it brings, it bridges the

gap between people who want to help and people

who are in need of help.

How it works: It's very simple. Just four
steps.

As I said, the first step is, you look

at the various causes and various filter criterias
and

choose the person you need to help. You want

to help.

And make a loan on Milaap website. This is

just an online payment that you do. The real

test starts after website. So the whole business
is

about testing transferrency. Can you trust
Milaap that the

money is going in the right hands?

Another big thing, when it comes to helping
is,

donations - there is this hesitancy that,
you know,

my hard-earned money is going away, whereas
what you

do on Milaap is not a donation but a

loan. That means you help one person, let's
say

a person needs ten thousand rupees to build
a

toilet. You contribute just 500 rupees out
of it,

but that 500 is not a donation.

Within a year, month on month, you get back

your money, and all your 500 rupees back to

your account within one year's time. And then
you

can utilize the same thing to make an impact

on someone else's life. This is what we call

making an impact, again and again with the
same

money. And of course if you start believing
in

the idea, you start putting more, more of
money

in helping others, right.

So there's the four simple steps. Choose a
customer,

make a loan, get repaid and relend.

How can you help, right? Of course the obvious

thing is you know check out the website, if

you like the idea, like us on Facebook. Tweet

about us, write about us on your blog. We

also have a widget which you can put on

your blog saying that you support Milaap.

But if you want to do even more, right,

obviously you can actually be an active participant
on

our website. Go and make a loan to someone

there today. You could also do something interesting,
I

think one of the best gift ideas that you

could have is to give a Milaap gift card

to someone.

That is you are gifting someone the ability
to

help someone, and repeatedly keep doing it,
right.

Finally you could also start a campaign. If
you

want to go really big, you could adopt a

particular borrower on the website or adopt
a particular

cause, and decide to raise some funds to impact

that critical cause. So you can decide to
raise

some money for education. Let's say you're
running a

marathon, you can put out, we can put up

a page for you and help you raise some

of money through you, through that event.

But, right, I'm sure many of you would think,

this guy's coming here talking about some
company, Milaap

in three, four minutes, why should I go ahead

and do anything about it, right? So proof
of

the pudding is in it's eating, right.

So what we want you is to first get

real, first-hand experience of what Milaap
is and how

it helps people, how much you can trust it,

how transparant it is. So we are right now

giving away free credits, free Milaap credits
worth 500

rupees. Anyone who wants to try out what Milaap

is without spending a buck from your pocket
can

use this gift card, GCRC14, which some of,
same

as our twitter hash tag as well. And you'll

be able to make a loan on Milaap just

by using those 500 rupees.

You just click on lend button and when it

comes to actually making a payment, just write
in

the coupon and you'll be able to, you are

good to go, right.

Thanks a lot.

V.O.: Thanks Sarish?? [00:23:48].

SORU: Houston, can you copy me?

OK. So I don't have slides. So I'm Soru.

I'm nineteen. I'm still in college. So over
the

last summer I did a Google Summer of Code

with a ?? project [00:24:03]. And I built
this

GitHub for designers, so I thought I would
teach

that here so that you could know about it,

and I could, I saw that there are some

design-oriented start-ups so they could probably
give me feedback.

So the idea is that, like, when designers
collaborate,

especially opensource designers, they go to
mailing lists and

they share links, and so, let's say, Coby
decides

to make, make this banner. So he starts with,

he starts with the logo, and he puts it

on MGo?? [00:24:34] and shares a link on the

mailing list, and people go plus one, cool,
or

this is awesome.

And everybody gets that link. It's a pain
in

the ass. So like, this is kind of very,

it's like spam, it's also annoying, and there's
no

nice way to understand designer feedback.

So how it helps is, so you can, GlitterGallery

is not live right now. But I'll get it

up soon. So right now what you can do,

there is, you can go create a new design

project just like GitHub and you can create
SVG

files, or you can create, you can upload your

own SVGs or PNG or JPEG or whatever.

So after that, so let's say he created the

same thing inverted in color, and he can post

it there and he can share links with his

friends and anybody else on the app can log

in and he can comment over there, so. So

cool, plus one and all that.

So some comments make sense. Because these
comments are

nice, they make you feel good. But there are

certain comments which could, which could
say that, OK,

if you're going to project this, this won't
be

visible properly. So he can mark said comments
as

issues. They go to an issue tracker, and somebody

else can forward the project, so it becomes
theirs.

They can make changes to it and send back

poll requests to the original designer.

So he can, you know, pull back and. So

design projects can flourish that way. And
also we

have a feature called GlitterPosts. So something
like, I

don't know, maybe like medium, where you can
go

and explain what, what design process you
went through.

So every time you make a new change, anytime

you update, there is a commit happening in
the

background.

So we use the great library provided by GitHub.

They use the same thing. So every instance
has

a commit idea and all that. So you can

reference comments, so you can go there and
say

I started with this file, this is the idea

behind it, and now, next I did this, this

is how it went.

So it also serves as a nice way to

learn, also maybe a nice way to judge designers'

thought process. So, so I would just like
to

ask if any of you work with designers or

are designers yourselves, what could be some
better ways

to collaborate, and how else could I extend
this

project.

By the way, this is opensource. And next summer

for Google Summer of Code we might be taking

around two people. So if there are any students

here who would like to contribute, you can
ping

me later. Thanks.

SPEAKER: Thanks, Qwinik, for the opportunity
to be here,

first of all, because they sent us here just

to, to sponsor and represent Qwiniks.

It happens that I land in this company, and

I start to work with them as a programmer,

and I have a lot of difficulties to, to

adopt my German mindset to the Indian mindset.
And

I'm still having difficulties. I'm still not
understanding things,

and I'm trying, deeply.

So, but I want to advise, or talk about,

what, what was the things that I've faced
here,

and first thing that I want to advise, and,

is that this is opensource work. Opensource
work means

that there is no one willing to teach you.

You have to teach by yourself. And the code

is there. Please open libraries and write
libraries. Not

write what is the code inside, and not what

is the documentation says, because documentation
can be outdated.

So with this I will advise that you need

to handle your own will and what is the

consequences. So you need to be able to know

the difference between what you want and what
the

project wants. This is for people that works
either

for projects, companies or by freelancers.

And you need to think about the consequences.
You

need to think about what is the, what is

the impact of this code, what is the maintenance

of this code, and what is the purpose of

this code. So think about that before writing
the

first line of code.

The second thing is setting up your priorities.
So

I found that it's somehow some people wants
to

rush to the finish, just to find a solution,

and most of the time without understanding
what the

problem is.

I spent two, three days thinking what the
real

problem is. What is the thing that you have

to solve, and then solve it. It's become,
it

will become easier in time. Mostly for maintainers.

They other one will be, be humble. Not be

humble in the streets saying, I'm not so,
or

not so, be humble with your code. So don't

try to solve big problems. Solve the small
problems.

So don't focus on the solution and the feature

that you want to generate. Focus in the method

that you're generate in. Making it consistent,
make it

very small.

I am try to go because there is a

lot of the - I will jump the career

of the programming to the culture one. I'm
sorry,

I will say. Clients are not gods. So you

are, you can be against them. It's allowed.
You're

allowed to be against them.

Meaning I, don't want to take this seriously,
but

meaning, meaning with this, you are allowed
to say

no, you are allowed to say scopes. Saying
we

are not allowed to do this. We need more

time or we need this resources that we don't

have. So please take in mind this because
this

is a huge impact for the company owners, for

the people in project managers.

Project managers are waiting, expecting something
from you, and

if you will not deliver on time, or what

they're expecting, you're breaking big rules.
It's not a

small thing. You're breaking big, big rules.

With this, I want to add, you need to

be predictable. Whatever you do make it predictable
for

the project manager, for the company owner,
for the

client. Meaning if you're doing something
that you know

that will be wrong, just tell him, it will

be wrong. But make it predictable. So don't
make

people wait for you, to see what is, what

it will happen. One minute more.

OK, I just jump down to, so I will

say after coming here, I'm just seeing the
huge

effort you are doing to work in opensource.
I

have seen sadly for me that some books are

more expensive than what juniors earn in one
month.

So that's brokes my heart, some way, because
I

am able to buy in Europe we're able to

buy books as something, without side effects.
And here

people that ?? [00:31:25], for them it's impossible.

I mean the book is two, is twice what

they earn a month. So with that, it is

impossible, I know that it's impossible to
learn. So

I really admire the courage that you have
to

come here. I know the price that you're paying

for coming here. I know the effort that you're

doing and I will like to say that, I

really admire that.

And yeah. My time is over. This guy has

an android with him, so I better run.

SPEAKER: OK, so I'll, I'll just go over this

and then see if I can-

So, what I'm going to do is I, the

time I have, I'll try to sow some doubts

in your mind, and then make you go back

and think about some of the things that you've

believed about.

So like I'll talk about my experience working
with

ActiveRecord and why I think it's a very bad

thing for building business applications.

So ActiveRecord is good when you're building
web applications,

shallow web applications that have a database
and some

rich UI on top of it. But when you

are trying to build business applications
that have a

lot of heavy logic, it's not a good, a

good thing to do.

I'll briefly talk about why it's not such
a

good thing. One of the most problematic things
about

ActiveRecord is it's so much coupled with
databases. So

when I am trying to solve a problem that

is very business logic heavy, I tend to put

things that are difficult to, positions that
are difficult

to reverse later.

The database schema is something that is very
difficult

to reverse. ActiveRecord makes you do that
upfront. Any

model that is tied to ActiveRecord, you can't
write

this without connecting to database and all
that. We

need to define schema up front.

Which I think is troublesome. So it kind of

makes it difficult to decouple the core business
logic

from the framework. Most of the stuff you're
writing

is tightly coupled with the framework. Something
new comes

up, you need to rewrite the entire thing.
There's

no, nothing that can you, you can take to

that new, new framework.

There other things, like, so whenever you
encounter situations

which were not thought of when the framework
was

designed, there are no points that allow you
to

extend. There's no easy way to customize the
ID

generator, logic. There's no easy way to incorporate
composite

keys. So the project I was working on,w e

needed to use mySQL partitions. So I can't
use

mySQL ID generation. I have to use auto composite

keys.

And there was no easy way of doing it.

There's a plugin that provides composite key
support, if

you go look at the implementation of that,
you'd

be scared to use it. So it's like, gorilla

patching - monkey patching to the extreme.
So it

kind of like highlights some of the problems.
How

the code to framework has been written.

And there are other features that I like and

other form mapping?? frameworks [00:34:29].
Like you load a

bunch of objects, do something on it, at the

end they all get persisted to the database.
It's

automatic, dirty checking transaction right
behind. There's something called

a persistence data utility. I create an order,
order

line items, all this stuff.

I create save and it saves everything. So
limited

support for things like that.

These features are very important when you're
trying to

follow these practices, domain-driven design,
so you need to

work in terms of aggragates, so you need to

decouple your persistence logic from your
busienss logic. All

that is very, very hard to do if you're

using ActiveRecord.

So there are some alternatives.

So datamapper is a framework, but I think
it's

become stagnant. There's a new version of
it called

Ruby?? [00:35:13]. I don't see a release for
that

as yet. So those are some alternatives that
you

can look at.

Or, yeah. Look at something in other languages.
So

these are based on my observations of working
with

ActiveRecord for the last five months, and
I'm comparing

it with eight years I've spent working with
Hibernate,

and this looks like a toy compared to what

that can do.

Thank you.

NIRANJAN PARANJAPE: So, a few really good
talks. Heavy

talks. I'm gonna keep this light.

So any - talk abuot continous learning.

But before going there I want to talk about

what happens if you don't learn continuously.

So we have been doing interviews for quite
some

time. And these are a few gems I have

heard while doing interviews.

So I asked a question: Which programming language
do

you like?

Imagine the answer - come on.

AUDIENCE: Ruby!

N.P.: No. I program in Rails.

Go read RubyonRails dot org. You'll find something
different.

Someone asked me, why shouldn't I use GOTO?

After all language supports it.

Do you know anything about cyclomatic complexity?

Do you know what other feature language supports
which

are really bad programming constructs and
you should generally

avoid?

Go read about that.

So I started the interview. I spoke about
like,

OK we have been hearing a lot about, you

should ask skinny controllers and fat models,
and there

was a huge buzz about it.

Then we went and said that, no we actually

need thin models.

So where to I write my business logic.

And I wrote a response. Helpers.

I asked someone, how do you scale your application?

I got a response: I don't have to, I

deploy in Cloud.

I asked someone, can you talked about GC?

Oh, GC, garbage compiler. Let's talk about
it.

I asked someone, why is encapsulation important?

I heard, because the book says data security
is

important.

And, last noted, but not the least - I

have heard this, actually.

I don't believe in learning in my personal
time,

because I believe in work-life balance. Now,
my answer

to that person is, this is the doctor's roadside.

Are you going to go to him, or are

you going to go to him?

You want your doctor to continuously learn
and understand

his medical practices, and essentially keep
up with the

latest technology. You are a software programmer.
You are

You cannot afford to fall behind the scene.
You

have to keep learning.

How do you keep learning? Lots of books are

out there. How many of you have read Pragmatic

Programmer's Journeyman to Masterman?

How many of you have read it twice?

This is a book you should read every year.

It's a brilliant book. Coding Honor. Clear
coding. Wonderful

books. Read different blogs. Learn different
language or paradigm,

because what you learn there, you can essentially
translate

it into your understanding of programming
itself, and maybe

bring some parts in your actual code.

But just reading these things are not important.
Understand

the concept behind is important. Where do
you understand

the concept?

Maybe you can go to, read c2 dot com.

How many people know that? What is c2 dot

com?

It's a first Wiki. And then maybe you can

go to c2 dot com. And maybe you can

go to c2 dot com again, because that's a

brilliant, brilliant site.

And most importantly, contribute to opensource.
We hear a

lot about hey, you should contribute to opensource
because,

you know, give back to the community, it's
fun.

But there is one major aspect of contributing
to

opensource that's just learning, self-learning.
And that is something

which I have done a lot.

How do you start? You start by reading code.

Read what other people are writing.

That's the first step in understanding how
to write

better code.

Essentially it allows you to pair with good
people,

across the globe, and get their feedback about
your

code. Talk to them about their design, their
understanding

of programming. So it's a wonderful tool.

Find a mentor if possible, and just, don't
just

stop there. You have to actually go and become

a mentor and teach someone what you have learned.

So yeah. That's it.